Professional Documents
Culture Documents
Neural Networks: Kernel Based Algorithms
Neural Networks: Kernel Based Algorithms
Price
Price
Size Size Size
• Output space: 0 1 y
wm1
m1
xm
0.5 (1,1)
0.5 1.0 φ1
(0,1) and (1,0)
• When mapped into the feature space < 1 , 2 > (hidden layer), C1 and C2
become linearly separable. So a linear classifier with 1(x) and 2(x) as inputs
can be used to solve the XOR problem.
NN for the XOR problem
|| x t1 ||2
1 (|| x t1 ||) e with t1 (1,1) and t2 (0,0)
|| x t2 ||2
2 (|| x t2 ||) e
|| x t3 ||2
3 (|| x t3 ||) e
x1
|| x t4 ||2
t1 -1
4 (|| x t4 ||) e
y
x2 t2 -1
+1
|| x t1 ||2 || x t2 ||2 with t3 (1,0) and t4 (0,1)
y e e 1
If y 0 then class 1 otherwise class 0
But all these 4 phi’s may not be required to get linear separability
. .. . .
N 1 N 2 ...... NN wN d N
{ } N
ij i , j 1 ( x j xi )
w x
1
w x
3rd ed. Haykin
3rd ed Haykin
11 12 ...... 1N w1 d1
.. w d
21 2
2
. .. . .
N 1 N 2 ...... NN wN d N
{ } N
ij i , j 1 ( x j xi )
w x
1
w x Inverse ???
Types of φ
• Multiquadrics:
(r) (r c )
2 2 1
c0
2
w1
w1 w3 w2
RBF ARCHITECTURE
x1
w1
1
x2
y
wm1
m1
xm
• One hidden layer with RBF activation functions
1... m1
• Output layer with linear activation function.
y w11 (|| x t1 ||) ... wm1 m1 (|| x tm1 ||)
|| x t || distance of x ( x1,..., xm ) from vector t
HIDDEN NEURON MODEL
x1
φ( || x - t||)
x2 t is called center
is called spread
center and spread are parameters
xm
Hidden Neurons
• Architecture:
• RBF networks have one single hidden layer.
• FFNN networks may have more hidden layers.
• Neuron Model:
• In RBF the neuron model of the hidden neurons is different from the one of the
output nodes.
• Typically in FFNN hidden and output neurons share a common neuron model.
• The hidden layer of RBF is non-linear, the output layer of RBF is linear.
• Hidden and output layers of FFNN are usually non-linear.
Comparison with multilayer NN
• Activation functions:
• The argument of activation function of each hidden neuron in a RBF
NN computes the Euclidean distance between input vector and the
center of that unit.
• The argument of the activation function of each hidden neuron in a
FFNN computes the inner product of input vector and the synaptic
weight vector of that neuron.
• Approximation:
• RBF NN using Gaussian functions construct local approximations to
non-linear I/O mapping.
• FF NN construct global approximations to non-linear I/O mapping.
Pseudo Inverse
• Left inverse
(ATA)-1AT . A = Inxn
Full Column Rank
Right Inverse
A. AT(AAT)-1 = Imxm
Full Row Rank